Discovering the Knowledge for Linguistic Information Aggregation

Knowledge can be in form of rules, patterns, decision trees, tables, networks, and so on. For linguistic information aggregation, both the inputs and outputs are expressed by linguistic variables. Using some aggregation operators, the knowledge about the mechanism for aggregating linguistic information can be interpreted as configuration of a group of weights. Other than the methodology using weights given in advance, the approach proposed attempts to estimating the weights from a set of alternatives given with some information about aggregated results. The approach is illustrated by an example of urban traffic situation judgment.